机器学习预测ACEI的比较与分析

发布时间：2018-11-26 18:21

【摘要】：血管紧张素转换酶抑制剂(ACEI)对高血压的治疗具有重要意义。基于从结构复杂的化合物数据库中构建的候选小分子数据集,采用分子对接技术从数据集中筛选出样本构建分类模型。分别采用支持向量机、K近邻、决策树、随机森林和贝叶斯方法建立血管紧张素转换酶潜在抑制剂和非抑制剂的分类模型。经结果对比,支持向量机相比于其他方法有更高的预测率,其中模型总体预测率和相关系数分别为82.4%和0.653。研究表明,支持向量机方法对于虚拟筛选血管紧张素转换酶抑制剂具有良好的效果。
[Abstract]:Angiotensin converting enzyme inhibitor (ACEI) plays an important role in the treatment of hypertension. Based on candidate small molecule data set constructed from complex compound database, molecular docking technique was used to screen samples from the data set to construct classification model. Support vector machine (SVM), K-nearest neighbor, decision tree, stochastic forest and Bayesian methods were used to establish the classification models of angiotensin converting enzyme potential inhibitors and non-inhibitors, respectively. Compared with other methods, support vector machine has a higher prediction rate, in which the overall prediction rate and correlation coefficient of the model are 82.4% and 0.653 respectively. The results show that the support vector machine method has good effect on virtual screening of angiotensin converting enzyme inhibitors.
【作者单位】： 江南大学物联网工程学院;江南大学数字媒体学院;江南大学工业生物技术教育部重点实验室;
【基金】：国家自然科学基金(No.21541006)
【分类号】：R544.1;TP181
，

本文编号：2359283